Misoprostol for intrauterine device placement

Objectives

This is a protocol for a Cochrane Review (intervention). The objectives are as follows:

To examine the effectiveness of misoprostol for routine intrauterine device placement, on patient (e.g. pain, side effects, satisfaction) and provider outcomes (e.g. ease of placement, need for adjunctive placement measures, placement success), compared with placebo or no treatment.

Background

Description of the condition

Intrauterine devices (IUDs) provide high contraceptive effectiveness, user satisfaction, and continuation rates (Peipert 2011; Trussell 2018). IUDs are generally safe for women to use, including adolescents and nulliparous women, based on the World Health Organization (WHO) medical eligibility criteria for contraceptive use (WHO 2015), and the US medical eligibility criteria for contraceptive use (Curtis 2016). Worldwide, IUDs are the most commonly used reversible method of contraception (Buhling 2014a; Dean 2018), although IUD use ranges from less than 2% in some countries, to higher than 40% in other countries (Buhling 2014a). IUD use in the USA has increased in recent years. Based on data from the National Survey of Family Growth, current IUD use among American women of reproductive age was 3.5% in 2006 to 2010 (women aged 15 to 44 years), 7.9% in 2015 to 2017 (women aged 15 to 49 years), and 8.4% in 2017 to 2019 (women aged 15 to 49 years (Daniels 2018; Daniels 2020; Jones 2012)). Prevalence of IUD use in the USA in 2017 to 2019 remained lower than female sterilization (18.1%) and oral contraceptive pills (14.0%), but was the same as the use of male condoms (8.4% (Daniels 2020)). 

While first attempts at placing IUDs are usually successful, some placements may be painful for the patient, or technically difficult for the provider (Teal 2015). During an IUD placement, pain can occur at different steps of the procedure: application of the instrument to the cervix to stabilize the uterus, uterine sounding, IUD insertion through the cervix, and IUD deployment in the uterus. Patients have different pain thresholds and may experience different levels of pain during IUD placement. Risk factors for more pain with in‐office transcervical gynecological procedures, including IUD placement, have been described and include being nulliparous, being postmenopausal, having a history of dysmenorrhea, suffering from anxiety, and anticipating pain (Chaves 2021; Dina 2018; Hunter 2020; Ireland 2016). The potential for, or fear of pain with IUD placement has been reported as a barrier to IUD initiation and a reason why patients have chosen other contraceptive methods (Gomez 2015; Hubacher 2015; Potter 2014).

Another barrier to IUD use is provider concern about potential technical difficulty with placement, particularly for certain populations, including adolescents and nulliparous women (Buhling 2014b; Daniele 2017; Elkhateeb 2020). However, one study found that IUDs can be placed in nulliparous adolescents and young women aged 18 to 24 years with high (> 95%) and similar success rates as their parous counterparts without the use of adjunctive measures, by both physicians and advanced practice clinicians (Teal 2015). Another study found no difference in IUD placement success rates between nulliparous and parous women (mean age of 23 years) who received an IUD for emergency contraception without the use of adjunctive measures, placed by nurse practitioners in a community setting; although IUD placement success rates for both nulliparous (80%) and parous women (86%) were lower than expected (Dermish 2013). However, after participating in a training program that focused on managing difficult IUD placements, clinicians’ rate of successful IUD placements improved significantly (> 95% for both nulliparous and parous women); the program included a two‐hour didactic session by a family planning expert, hands‐on training, and proctored IUD placements with cervical anesthesia or dilation (Dermish 2016). Another study found differences in IUD placement success rates between nulliparous women (89%), women with prior cesarean delivery only (89%), and parous women (98%), among a sample of women having an IUD placed by a physician in the primary care setting; 61% were over 35 years old, and 64% used local anesthesia (Harvey 2012). This study also found that practitioners who placed fewer than 100 IUDs over the 12‐month study period, more frequently rated IUD placements as difficult (Harvey 2012).

Description of the intervention

Misoprostol is a synthetic prostaglandin E₁ analog developed for the prevention and treatment of peptic ulcers (Monk 1987). It is also commonly used for obstetric and gynecologic indications, including induction of labor, prevention and treatment of postpartum hemorrhage, management of incomplete and spontaneous abortion, and termination of pregnancy (Tang 2013). Misoprostol causes cervical softening, dilation, and uterine contractions (Allen 2009). Misoprostol tablets were developed to be taken orally, but other routes of administration are sublingual, buccal, vaginal, and rectal (Tang 2007). Common side effects of misoprostol include nausea, vomiting, diarrhea, abdominal pain, chills, shivering, and fever (Goldberg 2001).

How the intervention might work

The cervix must be successfully navigated to place an IUD. The cervical softening and dilation effects of misoprostol might make placement of an IUD less painful for patients and technically easier for the provider.  

Why it is important to do this review

Identifying effective approaches to reduce patient pain and improve provider ease of IUD placement may reduce barriers to IUD access and increase patient access to comprehensive contraceptive options, a key strategy to promote reproductive autonomy and equitable contraceptive care (Holt 2020). The information may be used by guideline groups that publish evidence‐based contraception guidance, such as the WHO, the US Centers for Disease Control and Prevention, the UK Faculty of Sexual and Reproductive Healthcare, and the Society of Obstetricians and Gynaecologists of Canada. 

Objectives

To examine the effectiveness of misoprostol for routine intrauterine device placement, on patient (e.g. pain, side effects, satisfaction) and provider outcomes (e.g. ease of placement, need for adjunctive placement measures, placement success), compared with placebo or no treatment.

Methods

Criteria for considering studies for this review

Types of studies

We will include parallel randomized controlled trials (RCTs), including those randomized at the individual or cluster level. 

We will exclude cross‐over trials, because this is not feasible for studies of the intervention evaluated in this review. We will also exclude non‐randomized trials, observational studies, cross‐sectional studies, case series, review articles, editorials, letters, and conference abstracts. 

We will include peer‐reviewed articles published in any language.

Types of participants

Participants of interest will be patients receiving an interval intrauterine device (IUD, i.e. placement outside the postabortion or postpartum period). We will include studies that examine placement of currently available levonorgestrel (LNG)‐releasing IUDs, or any copper T IUD, for women of any age, of any parity, and for any indication. We will include studies that examine multiple IUD types if most participants received an IUD that met the above‐mentioned criteria. We will include studies that examine mixed placement timing (interval and postpartum or postabortion) if findings are stratified by placement timing. 

We will exclude studies of placements occurring exclusively during the postpartum or postabortion period (defined for this review as up to six weeks postpartum and four weeks postabortion). 

Types of interventions

We will include studies that directly compare misoprostol with placebo or no treatment. We will include all misoprostol routes of administration, doses, and regimens, with no limits, as long as the misoprostol is given for IUD placement. 

Types of outcome measures

Primary outcomes

• Pain experienced with IUD placement (as rated by a visual analog scale or other tool) at specific time points (i.e. during tenaculum placement, during IUD placement, and highest level of pain after placement and before clinic discharge)

• Provider ease of placement (as rated by a visual analog scale or other tool)

• Need for adjunctive placement measures (e.g. cervical dilation, ultrasound guidance, local anesthesia, analgesia)

• Placement success

Secondary outcomes

• Patient satisfaction with procedure (as rated by a Likert scale or other tool, before clinic discharge)

• Misoprostol side effects occurring before clinic discharge (each of the following will be reported: nausea, vomiting, pre‐placement abdominal pain, cramping, diarrhea)

• Adverse events occurring before clinic discharge (each of the following will be reported: uterine perforation, vasovagal reaction, cervical laceration)

Search methods for identification of studies

The Fertility Regulation Group Information Specialist will conduct a search for all published, unpublished, and ongoing studies, without restrictions on language or publication status. The search strategies for each database will be modelled on the search strategy designed for MEDLINE ALL Ovid (Appendix 1).

Electronic searches

We will search the following databases from their inception: 

• Cochrane Central Register of Controlled Trials (CENTRAL) Ovid EBM Reviews

• MEDLINE ALL Ovid (1946 onwards)

• Embase.com (1974 onwards)

• World Health Organization Global Index Medicus

• Global Health Ovid (1973 onwards)

• Scopus

We will search the following trials registries:

• The World Health Organization International Clinical Trials Registry Platform (www.who.int/trialsearch)

• ClinicalTrials.gov (www.clinicaltrials.gov)

Searching other resources

We will check the bibliographies of included studies and any relevant systematic reviews identified, for further references to relevant studies. We will contact experts and organizations in the field to obtain additional information on relevant studies. If necessary, we will contact authors of included studies for data clarification and further information. We will consider adverse effects described in included studies only.

Data collection and analysis

Selection of studies

We will download all titles and abstracts retrieved by electronic searching to a reference management database and remove duplicates. Three review authors (LZ, AN, KC) will independently screen titles and abstracts for inclusion. We will retrieve the full‐text study reports or publication, and two review authors (LZ, AN) will independently screen the full text, identify studies for inclusion, and identify and record reasons for exclusion of the ineligible studies. We will resolve any disagreement through discussion, or if required, we will consult a third review author (KC). We will list studies that initially appeared to meet the inclusion criteria but that we later excluded in the characteristics of excluded studies table. 

We will collate multiple reports of the same study, so that each study rather than each report is the unit of interest in the review. We will record the selection process in sufficient detail to complete a PRISMA flow diagram (Liberati 2009; Page 2021).

Data extraction and management

We will use a standard data collection form for study characteristics and outcome data; we will pilot the form on at least one study in the review. Two review authors (LZ, AN) will independently extract the study characteristics from the included studies. This may include: 

• Methods: study design, number of study centers and location, study setting, withdrawals, date of study, follow‐up, approach to adjustment for design effects or confounding;

• Participants: number, mean age, age range, parity, IUD type, inclusion criteria, exclusion criteria, other relevant characteristics;

• Intervention: dose, route, regimen, co‐interventions, comparison, compliance;

• Outcomes: events, means, relative effects, time points reported, adjusted effect estimates and information about the confounders and design effects accounted for, intra‐cluster correlations for studies with clustering;

• Notes: funding for trial, notable conflicts of interest of trial authors, ethical approval. 

Two review authors (LZ, AN) will independently extract outcome data from included studies. We will note in the characteristics of included studies table if outcome data were reported in an unusable way. We will resolve disagreements by consensus, or by involving a third review author (KC).

Assessment of risk of bias in included studies

Two review authors (LZ, AN) will independently assess the risk of bias for key outcomes defined in this protocol using the Cochrane RoB 2 tool (Sterne 2019), and criteria outlined in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022a). We will resolve any disagreements by discussion, or by involving another author (KC). Our effect of interest will be the effect of assignment, also known as the intention‐to‐treat. We will assess the following domains of bias, using a series of signaling questions:

• Bias arising from the randomization process;

• Bias due to deviations from intended interventions;

• Bias due to missing outcome data;

• Bias in measurement of the outcomes;

• Bias in selection of the reported result.

An additional domain of bias will be assessed for cluster‐randomized trials (i.e. bias arising from identification or recruitment of individual participants within clusters). We will use the variants of RoB 2 for cluster‐RCTs if we identify eligible trials with this study design. 

For each outcome, we will use the signaling questions to categorize each domain as low risk of bias, some concerns, or high risk of bias. We will record answers to the signaling questions using software (e.g. RoB 2 Excel tool, Covidence), and make them available in an online repository (Covidence). We will summarize the risk of bias judgments across different studies for each of the domains, for each prespecified outcome.

For each study, we will derive an overall judgment from the tool, as follows:

• Low risk of bias: we consider the study shows a low risk of bias;

• Some concerns: we expect a few concerns are associated with the study in at least one domain, but it does not warrant categorization as a study with a high risk of bias for any domain; 

• High risk of bias: we consider the study to be at high risk of bias in at least one domain; or we observed a few concerns with multiple domains in the study, such that these concerns significantly lower confidence in the study results.

We will not exclude studies on the grounds of their risk of bias, but will clearly report the risk of bias when presenting the results of the studies. When summarizing the evidence on intervention effects for different outcomes, we will consider the risk of bias for the studies that contributed to analyses for that outcome. We will conduct sensitivity analyses to assess whether estimated effects differ when high risk of bias studies are excluded from analyses. 

Measures of treatment effect

We will analyze dichotomous data as risk ratios (RR), odds ratios (OR), or both, with 95% confidence intervals (CI), and continuous data as mean difference (MD) or standardized mean difference (SMD) with 95% confidence intervals. We will report outcomes adjusted for confounders or design effects (e.g. clustering), and where possible, used for meta‐analysis. We will ensure that we enter data into the analysis with a consistent direction of effect (i.e. reversing the numeric coding of scales if needed). 

We will use SMDs when studies use different scales to measure the same outcomes, necessitating the standardization of study results to a uniform scale before they can be combined. The SMD expresses the size of the intervention effect in each study relative to the variability observed in that study, thus studies for which the difference in means is the same proportion of the standard deviation will have the same SMD, regardless of the actual scales used to make the measurements. To interpret the SMD, we will use the Cohen effect size rubric, where 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen 1988). If possible, we will express the study SMDs using a recognizable and standard metric used by some included studies, or use other strategies to aid interpretability, as outlined in Chapter 15 of the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2022a).

For studies reporting results that are not provided in a format that can be directly entered into meta‐analysis, we will use guidance provided in Chapter 6 of the Cochrane Handbook for Systematic Reviews of Interventions to convert the data to the necessary format (Higgins 2022b).

Unit of analysis issues

We will perform the primary analysis per individual randomized. We will abstract information on the study design and unit of analysis for each study, indicating whether clustering of observations is present, due to allocation to the intervention at the group level or clustering of individually randomized observations (e.g. patients within clinics). We will abstract available statistical information needed to account for the implications of clustering on the estimation of outcome variances, such as design effects or intra‐cluster correlations, and whether the study adjusted results for the correlations in the data. When the study does not account for clustering, we will ensure that appropriate adjustments are made to the effective sample size following guidance provided in Chapter 6 of the Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2022b). Where possible, we will derive the intra‐cluster correlation for these adjustments from the trial itself, or from a similar trial. If an appropriate intra‐cluster correlation is unavailable, we will conduct sensitivity analyses to investigate the potential effect of clustering, by imputing a range of values of intra‐cluster correlation.

If any trials have multiple arms that are compared against the same control condition, and we need to include them in the same meta‐analysis, we will divide the control group numerators and denominators by the number of interventions to be included in the meta‐analysis, to avoid double counting observations. 

Dealing with missing data

If included studies have missing data for key study characteristics or key outcomes, we will contact investigators or study sponsors to attempt to obtain the missing information. 

We will calculate missing standard deviations or other necessary data using other data from the trial, such as confidence intervals, based on methods outlined in Chapter 6 of the Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2022b).

We will report the number of studies that have results missing for the synthesis of each outcome.

We will show all responses and data provided in the characteristics of included studies table. When we make any assumptions about missing data, we will report the potential impact in the Discussion section of the review.

Assessment of heterogeneity

We will describe the clinical diversity and methodological variability of the evidence in the review text, and will use tables to describe study characteristics, including design features, population characteristics, and intervention details. 

To assess statistical heterogeneity, we will visually inspect forest plots and describe the direction and magnitude of effects, and the degree of overlap between confidence intervals. We will also consider the statistics generated in forest plots that measure statistical heterogeneity. We will use the I² statistic to quantify inconsistency among the trials in each analysis. We will also consider the P value from the Chi² test to assess whether this heterogeneity is significant (P < 0.1). If we identify substantial heterogeneity, we will report the finding and explore possible explanatory factors, using prespecified subgroup analysis. 

We will use a rough guideline to interpret the I² value rather than a simple threshold, and will bear in mind that estimate measures of heterogeneity (I² and Tau²) are of high uncertainty when the number of studies is small (Deeks 2022).

• 0% to 40%: heterogeneity might not be important

• 30% to 60%: may represent moderate heterogeneity*

• 50% to 90%: may represent substantial heterogeneity*

• 75% to 100%: considerable heterogeneity*

*The importance of the observed value of I² depends on (1) the magnitude and direction of effects, and (2) the strength of evidence for heterogeneity (e.g. P value from the Chi² test, or a confidence interval for I²).

Assessment of reporting biases

If we have enough studies available for meta‐analysis to support a funnel plot (at least 10), we will create and visually inspect the funnel plot, and run a formal statistical test for asymmetry, as proposed by Egger and colleagues (Egger 1997). Data permitting, we will create a funnel plot for patient pain with IUD insertion, provider ease of placement, and placement success. If we observe funnel plot asymmetry, we will discuss whether this could be attributed to small study effects, rather than just a non‐reporting bias. If our review includes too few studies eligible for meta‐analysis, the ability to detect publication bias will be largely diminished, and we will simply note our inability to rule out possible publication bias or small study effects.

Data synthesis

We will undertake meta‐analyses to estimate pooled effects when the studies report adequate comparable data that can support statistical pooling. When we suspect that data are skewed, based on the reporting of median and interquartile ranges, we will note the skewness and discuss the implication, but will not pool medians with means. 

For outcomes with data that cannot be statistically pooled, we will present descriptive forest plots showing the individual study results to illustrate the range of effects reported. 

If data are adequate to support meta‐analysis, we will take an analytic approach based on an evaluation of the clinical and methodological diversity of the included studies, as well as the statistical heterogeneity. For rare outcomes, we will use the Peto odds ratio method. For more common outcomes, we will generate the pooled effect using the DerSimonian and Laird random‐effects estimation technique. We will consider calculating a fixed‐effect estimate using the Mantel‐Haenszel approach if we can assume that the included studies are estimating the same intervention effect, the intervention effects are relatively consistent in direction and magnitude, and heterogeneity is low (0% to 40%). We will also consider the Mantel‐Haenszel approach if there is evidence of potential variation in outcome effects by study size (i.e. small‐study effects). We will discuss the implications and assumptions of the choice of meta‐analysis model if results differ, but the default approach for this topic will be the random‐effects model. We will illustrate each meta‐analysis using a forest plot to display effect estimates and 95% confidence intervals for both individual studies' effects and the pooled effect. 

If we cannot summarize the study data quantitatively, we will follow guidance available for synthesis without meta‐analysis outlined in Chapter 12 of the Cochrane Handbook of Systematic Reviews of Interventions (McKenzie 2021), and guidance on the reporting of synthesis without meta‐analysis (SWiM) in systematic reviews (Campbell 2020).

Subgroup analysis and investigation of heterogeneity

We will interpret tests for subgroup differences in effects with caution, given the potential for confounding with other study characteristics and the observational nature of the comparisons, as recommended in Chapter 10.11.2 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2022). Subgroup analyses with fewer than five studies per category are unlikely to be adequate to ascertain valid difference in effects, and we will not highlight them in our results. When adequate data are available to conduct meaningful subgroup analyses, we will evaluate factors that could explain observed statistical heterogeneity. We will conduct a statistical test for interactions with either a simple significance test to investigate differences between two or more subgroups, or use meta‐regression to evaluate potential subgroups differences in outcomes according to the factors described below (Borenstein 2013). We will only use meta‐regression if there are more than 10 studies available for meta‐analysis.

We plan to carry out subgroup analyses that may contribute to heterogeneity in the effects of the intervention. This may include subgroup analyses comparing studies grouped by the following factors: 

• Nulliparous versus parous (or prior vaginal delivery versus no prior vaginal delivery) 

• Prior failed IUD placement attempt versus no prior failed IUD placement attempt

• Misoprostol dose (< 400 mcg versus 400 mcg, and > 400 mcg versus 400 mcg) 

• IUD type (smaller versus larger width of inserter, and smaller versus larger IUD size)

If studies do not report stratified results, results may be grouped into categories that reflect the majority of participants, or will include a subgroup category of mixed populations. If there are enough studies reporting the outcome to support valid subgroup comparisons, we will conduct subgroup analysis for the following outcomes: patient pain with IUD placement (during IUD placement or highest level of pain after placement and before clinic discharge [whichever measure is most consistently reported]), and provider ease of placement.

Sensitivity analysis

We will perform the following sensitivity analyses to assess the robustness of our conclusions. Given that there is no formal statistical test that can be used for sensitivity analysis, we will provide informal comparisons between the different ways of estimating the effect under different assumptions. We will report sensitivity analysis results in tables rather than forest plots.

• Restricting the analyses to studies with a low risk of bias, as described under assessment of risk of bias in included studies

• Restricting the analyses to studies with no clustering of participants, or clustering of participants that included an appropriate intra‐cluster correlation

Summary of findings and assessment of the certainty of the evidence

We will evaluate the certainty of the evidence for each of our prespecified outcomes according to the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias).

We will follow the methods and recommendations described in Chapter 14 of the Cochrane Handbook of Systematic Reviews of Interventions (Schünemann 2022b), and will use GRADEpro GDT software to develop summary of findings tables (GRADEpro GDT). 

In the summary of findings table(s), we will provide results for the following key outcomes: 

• Pain experienced with IUD placement 

• Provider ease of placement 

• Need for adjunctive placement measures 

• Placement success

• Patient satisfaction with procedure 

• Misoprostol side effects occurring before clinic discharge 

• Adverse events occurring before clinic discharge 

We will use footnotes to give justifications for our decisions to downgrade the certainty of evidence, and provide comments to aid readers’ understanding of the review when necessary. Two review authors (LZ, AN) will make independent judgments about the certainty of the evidence, with disagreements resolved by discussion or involving a third author (KC). We will justify the judgments, document them, and incorporate them into reporting of results for each outcome.

Appendices

Appendix 1. Model Search Strategy

MEDLINE ALL (Ovid) <1946 to July 05, 2022>

Date searched: 6 July 2022

1 intrauterine devices/ or intrauterine devices, medicated/ or intrauterine devices, copper/ (12022)

2 (IUB or IUBs or IUC or IUCs or IUD or IUDs or IUCD or IUCDs or IUS or IUSs or CuIUB or Cu‐IUB or CuIUBs or Cu‐IUBs or CuIUD or Cu‐IUD or CuIUDs or Cu‐IUDs or CuIUC or Cu‐IUC or CuIUCs or Cu‐IUCs or CuIUCD or Cu‐IUCD or CuIUDs or Cu‐IUDs or CuIUS or Cu‐IUS or CuIUSs or Cu‐IUSs or ECIUD* or LNGIUC or LNGIUCs or LNGIUCD or LNGIUCDs or LNGIUD or LNGIUDs or LNGIUS or LNGIUSs).ti,ab,kf. (11542)

3 ((intrauterine or intra‐uterine) adj3 (ball or balls or coil or coils or contraceptive or contraception or device or devices or system or systems)).ti,ab,kf. (10180)

4 (Kyleena or Liletta or Mirena or Skyla or Copper‐7 or Copper‐T or CuSafe or Cu‐Safe or Cu375 or Cu‐375 or "Cu 7" or "Cu T" or "Cu T‐200" or CuT200 or CuT380* or Cu‐T380* or FlexiT or Flexi‐T or FlexiT300 or Flexi‐T300 or Gyne or Gynefix or Gyneplus or Gyne‐T380S or Liberte or "Lippes Loop" or Load‐375 or MCu or MLCu* or "ML Cu375" or Mini380 or "Mini 380" or Mini‐TT or MiniTT or "Mona Lisa" or Multiload or Multi‐load or MultiSafe or Multi‐Safe or MYCu or NeoSafe or Neo‐Safe or NovaT or Nova‐T or NovaT380 or Paragard or TCu or TSafe or T‐Safe or T380* or T‐380* or TT380* or TT‐380* or UT380 or UT‐380).ti,ab,kf. (4499)

5 or/1‐4 (21179)

6 Misoprostol/ or Prostaglandins E, Synthetic/ (5911)

7 (misoprostol or PGE or PGE1 or prostaglandin or prostaglandinE1).ti,ab,kf. (93274)

8 (Asotec or Chromalux or Cyprostol or Cytil or Cytopan or Cytotec or Cyrux or Gastrul or Isovent or Kontrac or Misofar or Misoprolen or Misotac or Misotrol or Mizoprostol or Noprostol or Novo‐Misoprostol or Prostokos or Vagiprost).ti,ab,kf. (150)

9 or/6‐8 (94213)

10 and/5,9 (241)

11 10 not ((Animals/ not Humans/) or (baboon$1 or bovine or canine or capra or cattle or cow$1 or deer or dog$1 or equine or ewe$1 or hamster$1 or goat$1 or mares or mice or mouse or ovine or rabbit$1 or rat$1 or rattus or rodentia or sheep).ti.) (206)

Contributions of authors

Conceiving the protocol: LZ, AN, ES, NK, AT, MW, KC

Designing the protocol: LZ, AN, ES, NK, AT, MW, KC

Co‐ordinating the protocol: LZ, AN, ES, NK, AT, MW, KC

Designing search strategies: Robin Payter (Information Specialist)

Writing the protocol: LZ, AN, ES, NK, AT, MW, KC

Performing previous work that was the foundation of the current study: LZ, AN, ES, NK, AT, MW, KC

Sources of support

Internal sources

• none to declare, Other

  none

External sources

• none to declare, Other

  none

Declarations of interest

LZ: none known

AN: none known

ES: none known

NK: none known

AT: none known

MW: none known

KC: none known

New